railroad grade crossing accident analysis at microscopic … w. hay... · · 2018-02-27railroad...
TRANSCRIPT
![Page 1: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/1.jpg)
Sponsored by
William W. Hay Railroad Engineering Seminar
Railroad Grade Crossing Accident Analysis at Microscopic and Macroscopic levels
Ray BenekohalProfessor
University of Illinois at Urbana-Champaign
Date: Friday, November 14, 2014 Time: Seminar Begins 12:15
Location: Newmark Lab, Yeh Center, Room 2311University of Illinois at Urbana-Champaign
Juan MedinaPostdoctoral Research Associate
University of Illinoisat Urbana-Champaign
![Page 2: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/2.jpg)
RAILROAD GRADE CROSSING ACCIDENT ANALYSIS AT
MICROSCOPIC AND MACROSCOPIC LEVELS
Rahim F. BenekohalJuan C. Medina
Nov 14, 2014 – University of Illinois at Urbana-Champaign
William W. Hay Railroad Engineering Seminar
2
![Page 3: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/3.jpg)
1 – Introduction2 – Research Questions3 – A Micro Approach4 – State of Practice – The U.S. DOT Model5 – A Combined Model7 – Accuracy of Predictions / Rankings8 – Future Research
Outline
3
![Page 4: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/4.jpg)
Fatalities at Public Grade Crossings
Introduction
* Source: NCHRP 755
Grade crossing accidents are more likely to be more severe, more costly, and to involve a fatality than other highway crashes (NCHRP 755, 2013)
Fatal accidents have remained at ~10% of accidents for the last 25 years
4
![Page 5: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/5.jpg)
Accidents at Grade Crossings (2003 – 2012)
Introduction
Nationwide Data Illinois Data
Crossings remain a significant hazard. Latest trends do not show accident decline (NCHRP 755)
5
![Page 6: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/6.jpg)
Distribution of accidents in Illinois
Introduction
Similar general patterns, but difficult to predict specific locations due to low frequencies
2007 - 20082003 - 2004 2010 - 2011
6
![Page 7: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/7.jpg)
Distribution of accidents at public crossings in Illinois (2003 – 2012)
Introduction
FrequencyLocations in the analysis
%
0 5874 87.22%1 685 10.17%2 128 1.90%3 31 0.46%4 11 0.16%5 4 0.06%7 1 0.01%9 1 0.01%
Total 6735 100.00%
- Accidents occurred at ~13% of crossings
(similar nationwide trend ~12%)
- Highest accident frequency = 9
- Only 48 locations with 3 or more accidents
- Highly skewed distribution, as expected
7
![Page 8: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/8.jpg)
Can we find trends, contributing factors by analyzing individual high accident locations? (micro approach)
If there are trends, can we incorporate them in a macro model?
• In current models?• In new models?
If we can incorporate them in a model, can we improve:
• Accident predictions?• Ranking of locations?
Research Questions
8
![Page 9: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/9.jpg)
A Micro Approach – Single Crossing
• Micro-level approach is being developed
• The main goal is to spot local trends/contributing factors
• Simple, easy-to-use, and complements info for site inspections and prelim analysis
• 3 steps so far:
• Sketch of crossings with key info• Tree structure to spot trends• Additional information (e.g. surroundings, land use, nearby ramps)
9
![Page 10: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/10.jpg)
A Micro Approach – Single Crossing
• Sketch of crossings with key info:
Nagle Ave and Avondale Ave, Near I-90, Chicago
10
![Page 11: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/11.jpg)
A Micro Approach – Single Crossing
• Variables observed:
11
![Page 12: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/12.jpg)
A Micro Approach – Single Crossing
• Variables observed:
12
![Page 13: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/13.jpg)
A Micro Approach – Single Crossing
13
![Page 14: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/14.jpg)
A Second Example:
South Chicago 119 St and Wood St(near I-57)
- 7 accidents in 10 years
- All accidents on main tracks
A Micro Approach – Single Crossing
14
![Page 15: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/15.jpg)
A Micro Approach – Single Crossing
A Second Example:
No clear trends
15
![Page 16: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/16.jpg)
A Third Example:
- Chestnut and Lehigh Ave, Glenview, IL
- 4 of 5 accidents involved drivers older than 80
- The remaining driver was 61 years old
A Micro Approach – Single Crossing
16
![Page 17: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/17.jpg)
A Micro Approach – Single Crossing
A Third Example (Cont…)
- High concentration of assisted living communities
- Could countermeasures incorporate population type?
- Potential age-related solutions (signs, marking)?
17
![Page 18: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/18.jpg)
Let’s look at a common model to predict accidents at grade crossings…
FRA Web Accident Prediction System (WBAPS)
This system uses the DOT accident prediction formula (developed from 1975 database)
Accident predictions are an input to resource allocation model
State of Practice
18
![Page 19: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/19.jpg)
Presenter:
Dr. Juan C. MedinaPostdoctoral Research Associate
19
![Page 20: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/20.jpg)
The U.S. DOT Model
• Initial collision model was developed based on the data from 1975 database
• Most coefficients remain unchanged since 1980
• An accident history term was added to reflect safety trends in recent years
• Weighted average of the initial and accident history terms is computed
• The computed value is multiplied by “normalizing constants” that vary by year
20
![Page 21: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/21.jpg)
Three Main Steps in U.S. DOT Model
1. Initial collision prediction (a) = 𝐾𝐾 × 𝐸𝐸𝐸𝐸 × 𝑀𝑀𝑀𝑀 × 𝐷𝐷𝑀𝑀 × 𝐻𝐻𝐻𝐻 × 𝑀𝑀𝑀𝑀 × 𝐻𝐻𝑀𝑀 × 𝐻𝐻𝐻𝐻
Source: U.S. DOT and Federal Highway Administration. Railroad-highway Grade Crossing Handbook. Washington D.C., 2007
Example for Flashing Lights:
21
![Page 22: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/22.jpg)
Three Main Steps in U.S. DOT Model
2. Second collision prediction (B) =𝑀𝑀0
𝑀𝑀0 + 𝑀𝑀 𝑎𝑎 +𝑀𝑀
𝑀𝑀0 + 𝑀𝑀𝑁𝑁𝑀𝑀
Accident History Term(N accidents in T years)
𝑀𝑀0 =1.0
0.05 + 𝑎𝑎
3. Final collision prediction (A) = B * Normalizing Constant
Source: Accident Prediction and Resource Allocation Procedure Normalizing Constants. 2010
22
![Page 23: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/23.jpg)
Other Alternatives… A Combined Model?
• Consider findings from micro analysis in building a macro model
• Used data from Illinois, explored different regression models:• Poisson, NB, Zero-inflated NB
• 5 years of data (2003-2007) used for model building•
• 5 years for data (2008-2012) used for evaluation/validation of the model
• ZINB models created for the same warning devices used by U.S. DOT formula
23
![Page 24: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/24.jpg)
• ZINB Model Form:
𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = 1 −1
1 + 𝐴𝐴 �−(𝑘𝑘1− 𝑘𝑘2∗𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎∗ 𝐴𝐴−𝑘𝑘3+ 𝑘𝑘4∗𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 + 𝑘𝑘5∗𝑎𝑎𝑡𝑡𝑎𝑎𝑡𝑡𝑡𝑡𝑡𝑡 +(𝑘𝑘6∗𝑙𝑙𝑎𝑎𝑡𝑡𝑙𝑙𝑡𝑡)+𝑓𝑓𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎
𝑓𝑓𝑎𝑎𝑡𝑡𝑎𝑎𝑙𝑙𝑙𝑙 = �𝑘𝑘7 𝐴𝐴𝑓𝑓 0 < 𝑎𝑎𝐴𝐴𝑎𝑎𝑎𝑎𝐴𝐴 < 30𝑘𝑘8 𝐴𝐴𝑓𝑓 30 ≤ 𝑎𝑎𝐴𝐴𝑎𝑎𝑎𝑎𝐴𝐴 < 600 𝐴𝐴𝑓𝑓 60 ≤ 𝑎𝑎𝐴𝐴𝑎𝑎𝑎𝑎𝐴𝐴 ≤ 90
A Combined Model
24
![Page 25: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/25.jpg)
Selected for Final Model
Pvalue
Intercept - < 0.0001aadt - 0.0383
total_train - 0.00981 (0◦-29◦)2 (30◦-59◦)3 (60◦-90◦)
total_tracks - x N/Atraf_lanes - x N/A
1 (0-200ft)2 (>200ft)
Intercept (zero model) - 0.35 *total_train (zero model) - 0.18 *
N/A
Passive Warning Devices (e.g. crossbucks)
hwy_near
x
x
Variable Category
N/Across angle
A Combined Model• Model for crossbucks based only on conflicting traffic (similar to US DOT model)
25
![Page 26: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/26.jpg)
Selected for Final Model
Pvalue
Intercept - < 0.0001aadt - 0.0089
total_train - 0.02261 (0◦-29◦) 0.07462 (30◦-59◦) 0.06673 (60◦-90◦) -
total_tracks - x N/Atraf_lanes - 0.0044
1 (0-200ft)2 (>200ft)
Intercept (zero model) - 0.0438total_train (zero model) - 0.0275
N/A
Active Warning Devices (e.g. flashing lights)
hwy_near
x
Variable Category
cross angle
A Combined Model• Angle was a significant contributing factor for active warning devices
26
![Page 27: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/27.jpg)
Selected for Final Model
Pvalue
Intercept - < 0.0001aadt - < 0.0001
total_train - < 0.00011 (0◦-29◦)2 (30◦-59◦)3 (60◦-90◦)
total_tracks - 0.0003traf_lanes - x N/A
1 (0-200ft) 0.02832 (>200ft) -
Intercept (zero model) - 0.2521*total_train (zero model) - 0.2495*
N/Ax
Gates
hwy_near
Variable Category
cross angle
A Combined Model• Angle not important for gated crossings; distance to nearest highway
intersection was significant
27
![Page 28: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/28.jpg)
• Comparison is based on two factors:
1. Absolute and relative predictions
2. Ranking of high accident locations
Accuracy of Prediction
28
![Page 29: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/29.jpg)
Accuracy of Prediction
• (Ʃ Predicted)/(Total observed)
• True prediction of all crossings together
• Useful to check magnitude of predictions
Overall absolute predictions:
0
0.5
1
1.5
2
2.5
3
3.5
4
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Σ(Pr
edic
ted)
/Tot
al O
bser
ved
Cumulative Proportion of Crossings
Active Warning Devices
Field Data
US DOT Model
ZINB Model - Average Modeland Data
29
![Page 30: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/30.jpg)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Σ(Pr
edic
ted)
/Tot
al O
bser
ved
Cumulative Proportion of Crossings
Crossings with Gates
Field Data
US DOT Model
ZINB Model - Average Model and Data
Accuracy of Combined Macro Model
0
0.5
1
1.5
2
2.5
3
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Σ(Pr
edic
ted)
/Tot
al O
bser
ved
Cumulative Proportion of Crossings
Passive Warning Devices
Field Data
US DOT Model
ZINB Model - Average Model and Data
30
![Page 31: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/31.jpg)
• (Ʃ Predicted)/(Total Predicted)
• Useful to check if prediction curve is similar to data
Overall relative predictions:
0
0.2
0.4
0.6
0.8
1
1.2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1Σ(
Pred
icte
d)/T
otal
Pre
dict
ed
Cumulative Proportion of Crossings
Active Warning Devices
Field DataZINB ModelUS DOT Model
Accuracy of Prediction31
![Page 32: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/32.jpg)
0
0.2
0.4
0.6
0.8
1
1.2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Σ(Pr
edic
ted)
/Tot
al P
redi
cted
Cumulative Proportion of Crossings
Crossings with Gates
Field DataUS DOT ModelZINB Model
0
0.2
0.4
0.6
0.8
1
1.2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Σ(Pr
edic
ted)
/Tot
al P
redi
cted
Cumulative Proportion of Crossings
Passive Warning Devices
Field Data
ZINB Model
US DOT Model
Accuracy of Combined Macro Model
32
![Page 33: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/33.jpg)
Ranking of high accident locations
Top 10 Top 20Data (observed) 11 21US DOT Formula 6 10
ZINB - Average model and data 8 14Data (observed) 16 26US DOT Formula 8 13
ZINB - Average model and data 11 19Data (observed) 28 48US DOT Formula 18 31
ZINB - Average model and data 20 35
Active (Flashing Lights)
Gates
Ranking MethodWarning Device
Passive (Crossbucks)
Number of crashes predicted in top
locations
33
![Page 34: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/34.jpg)
Summary
• Improvements for state-of-practice in accident prediction are needed:• Accuracy of predictions, understanding contributing factors, finding countermeasures
• Micro approach finds contributing factors that data aggregation may mask•
• Combination of macro and micro analysis improved accident prediction
• Better accuracy• Better ranking
34
![Page 35: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/35.jpg)
Future (and Ongoing) Research
• Additional accident model forms, selection of best option
• Dynamic tree structure to automatically sort attributes and discover trends• Cluster crossings based on potential trends (corridors)
• Corridor Analysis
• Approaches for combining macro and micro
… More
35
![Page 36: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/36.jpg)
A Micro Approach – Extension to Corridors?
• Can we extend the micro approach to corridors?
• Accidents along corridor can be grouped to identity possible trends
• GIS to locate crossings, add data with socio-economic and geographic info
Future (and Ongoing) Research
36
![Page 37: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/37.jpg)
Example Corridor
• Northeast Illinois Regional CommuterRailroad
25 accidents at 8 crossings between 2003 and 2012
Future (and Ongoing) Research
37
![Page 38: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/38.jpg)
Combining Micro and Macro• One idea is to simply include variables from micro to macro models
• We could also add categories to the macro model (dummy variables, indicators, different models) and apply factors to modify predictions of some crossings
• Micro could also lead to different classification of crossings based on risk assessments
Future (and Ongoing) Research
38
![Page 39: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/39.jpg)
Future (and Ongoing) Research
...Continued
• Temporal analysis on the occurrence of accidents
• Verification with other datasets:• So far used Illinois data. How to generalize it
39
![Page 40: Railroad Grade Crossing Accident Analysis at Microscopic … W. Hay... · · 2018-02-27Railroad Grade Crossing Accident Analysis . at Microscopic and Macroscopic levels. Ray Benekohal](https://reader031.vdocuments.site/reader031/viewer/2022020411/5aba5b8a7f8b9ab1118bc0a7/html5/thumbnails/40.jpg)
Questions?
Rahim F. Benekohal ([email protected])Juan C. Medina ([email protected])
Nov 14, 2014 – University of Illinois at Urbana-Champaign
William W. Hay Railroad Engineering Seminar
40